UAV-DualCog - Self-Relative Position: leaderboard
Metric: Answer accuracy (%; 1,024 questions asking where the UAV is relative to a landmark from 2 or 5 aerial images; structured JSON answers on renders of simulated AerialVLN scenes, the discrete answer scored whatever the predicted box or interval; instant mode with explicit thinking disabled where a model allows it; higher is better). Source: arxiv.org. Saturation forecast: Around 2032. 36 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 (Non-reasoning) | 38 |
| 2 | GPT-5.4 (Non-reasoning) | 37.9 |
| 3 | Claude Sonnet 4.6 | 37.7 |
| 4 | Kimi K2.6 (Non-reasoning) | 36.1 |
| 5 | GPT-5.3 Instant | 35.2 |
| 6 | Kimi K2.5 (Non-reasoning) | 34.3 |
| 7 | Gemini 3.1 Flash Lite (Non-reasoning) | 33.7 |
| 8 | Qwen 3.7 Plus (Non-reasoning) | 33.3 |
| 9 | Qwen 3.6 Plus (Non-reasoning) | 32.8 |
| 10 | Qwen 3.5 122B A10B (Non-reasoning) | 32.6 |
| 11 | Qwen 3.5 27B (Non-reasoning) | 31.6 |
| 12 | Qwen 3.5 4B (Non-reasoning) | 30.9 |
| 13 | GLM-4.6V (Non-reasoning) | 30.8 |
| 14 | Qwen 3.5 35B A3B (Non-reasoning) | 29.5 |
| 15 | Qwen 3.5 9B (Non-reasoning) | 29 |
Interactive version: theaggregate.ai/benchmark?slug=uav-dualcog-self-relative-position · How It Works · Data refreshed daily, snapshot 2026-09-29.